Author's response to reviews Title: The Effects of Hyperbaric Oxygen Therapy on Oxidative Stress, Inflammation, and Symptoms in Children with Autism: an Open-label Pilot Study Authors: Daniel A Rossignol (rossignolmd@gmail.com) Lanier W Rossignol (dlross7@hotmail.com) Jill James (JamesJill@uams.edu) Stepan Melnyk (MelnykStepanB@uams.edu) Elizabeth Mumper (afc-em@ntelos.net) Version: 3 Date: 24 October 2007 Author's response to reviews: see over
Authors response to reviews: second letter Original title: The Effects of Hyperbaric Oxygen Therapy on Oxidative Stress and Inflammatory Markers in Autistic Children: a Pilot Study Updated title: The Effects of Hyperbaric Oxygen Therapy on Oxidative Stress, Inflammation, and Symptoms in Children with Autism: an Open-label Pilot Study Authors: Daniel A. Rossignol MD Lanier W. Rossignol FNP S. Jill James PhD Stepan Melnyk PhD Elizabeth Mumper MD
October 24, 2007 Dear BMC Editors and Referees: Re: MS: 1965641892143060 The Effects of Hyperbaric Oxygen Therapy on Oxidative Stress, Inflammation, and Symptoms in Children with Autism: an Open-label Pilot Study We have reviewed the remaining critiques from referees # 1 and # 3 and have provided our answers in blue (attached). We thank the referees for their additional comments in order to improve the manuscript quality and readability. The previous changes to the manuscript as submitted in blue and red have been made. Any new changes beyond this have once again been marked with blue for additions, and erased words are in red with a strikethrough. Based upon Referee #1 s comments, we have added a discussion about the potential use of oxygen by mask alone compared to the oxygen and pressure delivered in the 1.3 atm group. We have also addressed the remaining concerns by both referees. Once again, we are thankful for the opportunity to respond to these concerns. Some other general changes were made (again in blue). Since we were unsure if the clinical data would be published, we have now made some corrections to the manuscript that are mainly grammatical. The abstract was rewritten for clarity. 2 references were removed that were duplicates (# 20 and 41). Several references were added (22, 28, 68, 81, and 91). All statistics were re-analyzed using the SigmaStat program and several nonsignificant errors were uncovered (p value on SRS motivation went from 0.011 to 0.021, ABC-C irritability from 0.430 to 0.618, ATEC health/physical/behavior from 0.625 to 0.438, CRP at 1.3 atm from 0.100 to 0.123, CRP at 1.5 atm from 0.099 to 0.084, CRP for all 18 children from 0.087 to 0.021, ATEC total at 1.3 atm from 0.030 to 0.007, ABC-C total at 1.5 atm from 0.063 to 0.094). All other p- values were confirmed. We apologize for these errors which appear to be due to using an older and inferior statistical analysis program. Several typos in the manuscript were also corrected. We look forward to any other concerns that need to be addressed. Respectfully, Dan Rossignol MD Principal Investigator, on behalf of all authors
Referee 1: Reviewer's report Title: The Effects of Hyperbaric Oxygen Therapy on Oxidative Stress and Inflammatory Markers in Autistic Children: a Pilot Study Version: 2 Date: 17 September 2007 Reviewer: Barry Russman Reviewer's report: As reviewer one, the following comments are made: 1. The comment that in an opened-labeled study, a p-value of.001 will be necessary in order to publish clinical outcomes was an exaggeration. In a double-blind controlled study a p-value of <.05 is significant. However, in an opened-labeled study I would personally like to see more than a p-value of.05 to occur when comparing efficacy of an intervention when using pre and posttesting and not using a control group. 2. The authors also argue that function or behavioral changes are important to note in openlabeled studies. I agree. They also noted that Montgomery s article on the effects of hyperbaric oxygen therapy in children with spastic diplegic cerebral palsy: a pilot project, showed some efficacy. In fact, the study published by Collet, et al was developed as a result of Montgomery s initial observation in 1999. Therefore, I also agree with that point. General I still have some concerns. 1. The authors lump together five studies that they claim show efficacy of HBOT in the treatment of cerebral palsy. I feel that the authors have misled the readers with this statement. The Collet study concluded that HBOT treatment has no advantages over treatment with slightly pressurized air in children with cerebral palsy. The global improvement seen in both treatments during the course of the study might be related to the context of the intervention and the selection of very motivated parents. A possible effect of increased pressure cannot be ruled out because of the slight increase in PaO2; nevertheless, the same increase in blood oxygen can be reached by simple mask administration of 28% fractional inspiration oxygen without increased pressure: a treatment that needs to be assessed before it can be recommended. The Collet study contained 2 treatment groups without an adequate control (placebo) group. However, in both groups the improvements were described as important and striking. Because there was no true placebo group, the interpretations of this trial have generated some differences of opinion. However, the other 4 studies cited did show benefits with HBOT in individuals with CP, in some cases dramatic. Given this background, we do not feel that we are misleading anyone, as interested readers can pull the studies and examine them for themselves. However, it is clear that further study on the use of HBOT in CP is needed, along with a description of the mechanism of action of HBOT in individuals with CP. Therefore, we have added the following underlined statement and 1 additional reference [28] to page 5: Several studies, using HBOT at similar pressures, also demonstrated clinical improvements in some
patients with cerebral palsy (CP) [23-26] that is some cases was dramatic [27]; however, some researchers have questioned the results of these studies and have called for further controlled trials and a focus on defining the mechanism of action of HBOT in individuals with CP [28]. I feel that the authors should point out to the readers that 1.3 liters of pressure with 24% oxygen is equivalent to taking oxygen by mask. This would give, in my opinion, more credibility to the article rather than just assuming that HBOT is helpful in the treatment of children with autism. We thank Dr. Russman for pointing this out. However, we respectively disagree with this opinion. Certainly, if oxygen were the only variable present, then the amount of oxygen generated at 1.3 atm and 24% oxygen might be equal to that which could be administered by mask alone. However, there is another variable present in the hyperbaric treatment delivered which is pressure. Several studies demonstrate that pressure, in and of itself, might lead to some additional clinical changes that cannot be obtained by simply using oxygen by mask alone. Please consider the following: 1. Granowitz, E. V., E. J. Skulsky, et al. (2002). "Exposure to increased pressure or hyperbaric oxygen suppresses interferon-gamma secretion in whole blood cultures of healthy humans." Undersea Hyperb Med 29(3): 216-25. This study examines the effects of hyperoxia, increased atmospheric pressure, and hyperbaric oxygen on cytokine synthesis. Five healthy volunteers were exposed to 90 min of room air, 100% oxygen, 10.5% oxygen at 2 atm abs, or 100% oxygen at 2 atm abs (HBO2). All subjects were blinded and randomly exposed to each of the 4 conditions. Immediately before entering the chamber, immediately after exposure, and 3 and 24 h later, blood was drawn and stimulated ex vivo with phorbol myristate acetate (PMA) and phytohemagglutinin A (PHA). Since lymphocytes are the primary source of PMA/PHAinduced interferon-gamma (IFN-gamma), these results were expressed as IFN-gamma production per 10(6) lymphocytes. Following the HBO2 exposure, PMA/PHA-stimulated lymphocytes released 51% less IFN-gamma than cells obtained before the exposure. This suppression persisted for 24 h following HBO2 (P < 0.05). Surprisingly, increased atmospheric pressure alone also inhibited IFN-gamma secretion (P < 0.05). Room air and hyperoxia alone had no significant effect upon IFN-gamma release. HBO2's anti-inflammatory effect may, in part, be due to inhibition of IFN-gamma release. 2. Shiratsuch, H. and M. D. Basson (2005). "Differential regulation of monocyte/macrophage cytokine production by pressure." Am J Surg 190(5): 757-62. BACKGROUND: Cytokine production by macrophages is essential for the inflammatory response. Normal human interstitial tissue pressure is 20 to 30 mm Hg, but generally decreases in acute inflammation. METHODS: We compared the effect of 20 mm Hg increased pressure (approximating normal interstitial tissue pressure) with that of ambient pressure (resembling pressure in inflamed tissues) on tumor necrosis factor (TNF)-alpha and interleukin (IL)-1beta production by undifferentiated (monocytic) and PMA (phorbol 12-, myristate 13-acetate)-differentiated (macrophage-like) THP-1 cells with or without lipopolysaccharide (LPS) (10 ng/ml). RESULTS: Pressure stimulated spontaneous macrophage TNF-alpha secretion (30.5 +/- 6.3 vs. 49.1 +/- 2.8 pg/ml, P <.02), but not monocyte TNF-alpha secretion. Pressure did not stimulate IL-1beta release.
As expected, LPS increased basal cytokine release. After LPS stimulation, pressure still tended to stimulate macrophage TNF-alpha, but inhibited monocyte TNF-alpha secretion (P <.05). In contrast, pressure inhibited IL-1beta release by both LPS-treated monocytes (986 +/- 134 vs. 595 +/- 226 pg/ml, P <.02) and macrophages (3,112 +/- 229 vs. 979 +/- 61 pg/ml, P <.01). CONCLUSIONS: Extracellular pressure may regulate TNF-alpha and IL-1beta secretion differentially by monocytes and macrophages. However, we do agree with Dr Russman that this topic needs to be discussed further in the paper. Therefore, we have added the following text to the discussion section starting on page 15: The amount of oxygen delivered in the 1.3 atm group is similar to the amount used in a previous study on HBOT in children with CP that utilized 1.3 atm and room air pressure ( hyperbaric air ) [24]. In that study, the authors commented that the amount of oxygen delivered at 1.3 atm was achievable with the use of 28% oxygen with a mask, without pressure ; however, this opinion did not account for the potential clinical effects of the increased atmospheric pressure delivered, because even low amounts of increased pressure may lead to significant clinical changes [44, 81]. Furthermore, the authors argued that hyperbaric air at 1.3 atm was unlikely to provide clinical benefit(s) because the mechanism of action of HBOT in CP is thought to be due to the penumbra phenomenon and that a clinical effect due to a pure pressure effect did not correspond to the rationale behind the hyperbaric oxygen treatment [24]. Since the mechanism of action of HBOT in autism may be different than in CP [32], including decreasing inflammation (as reviewed in the background section of this paper), it is entirely possible that clinical benefits may arise from purely increasing the atmosphere pressure delivered, because increased pressure delivery without additional oxygen appears to decrease inflammation (as measured by an inhibition of interferon-gamma release), and delivery of oxygen by mask without any increase in pressure may actually increase inflammation (as measured by an increase in interferon-gamma release) [44]. Since HBOT consists of 2 independent variables (pressure and inspired oxygen concentration), comparison studies are needed in individuals with autism before determining that the clinical effects of 1.3 atm and 24% oxygen are similar to those obtained by delivering oxygen by mask alone without additional pressure. In addition, further studies are needed that evaluate not only the clinical effects of hyperoxia delivered by HBOT, but also the effects of increased atmospheric pressure, because each of these effects may lead to different clinical outcomes depending on the underlying disease pathophysiology. 2. I suggest that the authors review the paper published by the Agency for Health Care Research and Quality, 2003. This is an evidence-based report on the use of HBOT therapy for brain injury, cerebral palsy, multiple sclerosis and stroke. The report concludes that the evidence from wellconducted clinical studies is limited. The balance of benefits and harms of HBOT for brain injury, cerebral palsy, or stroke has not been adequately studied. Future research of HBOT should include dose-ranging and safety studies to establish the optimum course of HBOT to evaluate in outcome studies. Future clinical trials should include several treatment options and should evaluate and measure caregiver burden in addition to patients functional outcomes. It would be helpful to this reviewer if the authors pointed out in their introduction that the use of HBOT in conditions other than the bends and burns is highly controversial and based mostly on anecdotal reports.
We are familiar with the 2003 AHRQ report and are aware that Dr. Russman helped to author this report. Since the use of HBOT in neurological conditions is a rapidly expanding field, we had the opportunity to review some additional articles in our paper that were not available to writers of that report, including some newer articles on HBOT and brain injury/neurological conditions, as well as articles on the effects of HBOT on inflammation and oxidative stress (which are particularly relevant to autism). These include references # 16-18, 21, 22, 25, 30, 32, 42, 44, 45, 48, 60, 61, 64-66, 81, 83, 85, and 92. These references have added to the growing body of evidence which supports the use of HBOT for neurological conditions. Furthermore, references # 12-14, 22, and 24 are controlled studies on the use of HBOT in several different neurological conditions; this documentation is certainly not anecdotal. In addition, since 2003, important case reports on chronic brain injury [18] and fetal alcohol syndrome [21] have been published. It is interesting to note that there are no published controlled studies on the use of HBOT in the bends or many of the other approved indications, but that these indications are accepted based upon clinical observations made over time. 3. I have to disagree with referee number two that, this is an important study of the efficacy and safety in autistic children. I feel that this is a safety report but certainly not an efficacy report. As noted by Collet in their article, the motivation of parents and the excitement of being involved in a new approach to the treatment of autism might be the explanation for the change in efficacy scores. We have added the following underlined/bolded text to page 20: The measurements of these clinical outcomes did have some inherent limitations and weaknesses. The use of parent-rated scales and the fact that parents were not blinded to the type of therapy given to their child might have introduced some bias. Furthermore, there was no placebo or control group. Therefore, the improvements found in this open-label study could be due merely to chance or to the natural development of the children. In addition, it is possible that any clinical improvements observed could have occurred as a result of the increased close interaction between the child and parent/caregiver, or motivation and/or enthusiasm that may have developed in the parent/caregiver during the course of the treatments. Because this was a pilot study, the sample sizes were small which makes it difficult to make adequate and meaningful comparisons between the 2 different pressures and oxygen concentrations used. Due to these issues, a larger double-blind, prospective study that includes a control group and more objective outcome measures is warranted. 4. To summarize: the report can discus safety issues about the use of HBOT in children with Autism, but more effort should be put into discussing why the children might have shown behavioral changes both to the pressurized oxygen and to the minimally pressurized oxygen (which is the same as giving oxygen by mask) such as parent enthusiasm, etc; similar to the caveats used by Collet, et al. Major Compulsory Revisions (that the author must respond to before a decision on publication can be reached) Minor Essential Revisions (such as missing labels on figures, or the wrong use of a term, which the author can be trusted to correct)
Discretionary Revisions (which the author can choose to ignore) What next?: Unable to decide on acceptance or rejection until the authors have responded to the major compulsory revisions Level of interest: An article of limited interest Quality of written English: Acceptable Statistical review: No, the manuscript does not need to be seen by a statistician. Declaration of competing interests: I have no competing interests. Barry S Russman
Referee 2: Reviewer's report Title: The Effects of Hyperbaric Oxygen Therapy on Oxidative Stress and Inflammatory Markers in Autistic Children: a Pilot Study Version: 2 Date: 9 October 2007 Reviewer: Philip James Reviewer's report: General It is important that this paper in covering the results of a preliminary study should be published as it should lead to controlled studies in this important field. Major Compulsory Revisions (that the author must respond to before a decision on publication can be reached) None Minor Essential Revisions (such as missing labels on figures, or the wrong use of a term, which the author can be trusted to correct) None, the revisions are acceptable. Discretionary Revisions (which the author can choose to ignore) What next?: Accept without revision Level of interest: An article of importance in its field Quality of written English: Acceptable Statistical review: Yes, but I do not feel adequately qualified to assess the statistics. Declaration of competing interests: I declare that I have no competing interests. We thank Dr. James for his comments and suggestions.
Referee 3: Reviewer's report Title: The Effects of Hyperbaric Oxygen Therapy on Oxidative Stress and Inflammatory Markers in Autistic Children: a Pilot Study Version: 2 Date: 18 September 2007 Reviewer: Stephen Lawrie Reviewer's report: General I reviewed this paper for BMC Med and the authors have responded well to my original criticisms. Major Compulsory Revisions (that the author must respond to before a decision on publication can be reached) None Minor Essential Revisions (such as missing labels on figures, or the wrong use of a term, which the author can be trusted to correct) I would however still like them to openly state in the paper why the group sizes were imbalanced (as they have done in their response to referees). We have added an explanation to page 8 about the 2 group sizes: This unequal division of children among the sample groups occurred due to scheduling constraints and because one center (EM) only treated the 1.3 atm group (6 children) while the other center (DR) treated both the 1.3 atm (6 children) and the 1.5 atm (6 children) groups. Discretionary Revisions (which the author can choose to ignore) None What next?: Accept after minor essential revisions Level of interest: An article of importance in its field Quality of written English: Acceptable Statistical review: Yes, and I have assessed the statistics in my report. Declaration of competing interests: I declare that I have no competing interests